CN116913508B - Method and system for predicting diabetic nephropathy based on white eye characteristics - Google Patents
Method and system for predicting diabetic nephropathy based on white eye characteristics Download PDFInfo
- Publication number
- CN116913508B CN116913508B CN202311174475.8A CN202311174475A CN116913508B CN 116913508 B CN116913508 B CN 116913508B CN 202311174475 A CN202311174475 A CN 202311174475A CN 116913508 B CN116913508 B CN 116913508B
- Authority
- CN
- China
- Prior art keywords
- data
- diabetic nephropathy
- training
- white eye
- white
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 208000007342 Diabetic Nephropathies Diseases 0.000 title claims abstract description 46
- 208000033679 diabetic kidney disease Diseases 0.000 title claims abstract description 45
- 241001469893 Oxyzygonectes dovii Species 0.000 title claims abstract description 42
- 238000000034 method Methods 0.000 title claims abstract description 33
- 238000012549 training Methods 0.000 claims abstract description 42
- 238000012216 screening Methods 0.000 claims abstract description 14
- 238000007477 logistic regression Methods 0.000 claims abstract description 9
- 239000008280 blood Substances 0.000 claims description 18
- 210000004369 blood Anatomy 0.000 claims description 18
- 210000004204 blood vessel Anatomy 0.000 claims description 18
- 206010012601 diabetes mellitus Diseases 0.000 claims description 14
- 210000003734 kidney Anatomy 0.000 claims description 12
- 102000008946 Fibrinogen Human genes 0.000 claims description 11
- 108010049003 Fibrinogen Proteins 0.000 claims description 11
- 229940012952 fibrinogen Drugs 0.000 claims description 11
- 102000001554 Hemoglobins Human genes 0.000 claims description 10
- 108010054147 Hemoglobins Proteins 0.000 claims description 10
- 201000010099 disease Diseases 0.000 claims description 8
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 claims description 8
- HVYWMOMLDIMFJA-DPAQBDIFSA-N cholesterol Chemical compound C1C=C2C[C@@H](O)CC[C@]2(C)[C@@H]2[C@@H]1[C@@H]1CC[C@H]([C@H](C)CCCC(C)C)[C@@]1(C)CC2 HVYWMOMLDIMFJA-DPAQBDIFSA-N 0.000 claims description 6
- DDRJAANPRJIHGJ-UHFFFAOYSA-N creatinine Chemical compound CN1CC(=O)NC1=N DDRJAANPRJIHGJ-UHFFFAOYSA-N 0.000 claims description 6
- 210000004185 liver Anatomy 0.000 claims description 6
- 238000007781 pre-processing Methods 0.000 claims description 6
- 210000000952 spleen Anatomy 0.000 claims description 6
- 230000007717 exclusion Effects 0.000 claims description 5
- 210000000813 small intestine Anatomy 0.000 claims description 5
- 210000002784 stomach Anatomy 0.000 claims description 5
- 238000012795 verification Methods 0.000 claims description 5
- 102000015779 HDL Lipoproteins Human genes 0.000 claims description 4
- 108010010234 HDL Lipoproteins Proteins 0.000 claims description 4
- 230000004872 arterial blood pressure Effects 0.000 claims description 4
- 238000001514 detection method Methods 0.000 claims description 4
- 210000002429 large intestine Anatomy 0.000 claims description 4
- 210000004072 lung Anatomy 0.000 claims description 4
- 102000004169 proteins and genes Human genes 0.000 claims description 4
- 108090000623 proteins and genes Proteins 0.000 claims description 4
- 238000011002 quantification Methods 0.000 claims description 4
- 210000002700 urine Anatomy 0.000 claims description 4
- 208000035473 Communicable disease Diseases 0.000 claims description 3
- WQZGKKKJIJFFOK-GASJEMHNSA-N Glucose Natural products OC[C@H]1OC(O)[C@H](O)[C@@H](O)[C@@H]1O WQZGKKKJIJFFOK-GASJEMHNSA-N 0.000 claims description 3
- 102000017011 Glycated Hemoglobin A Human genes 0.000 claims description 3
- 108010014663 Glycated Hemoglobin A Proteins 0.000 claims description 3
- 102000007330 LDL Lipoproteins Human genes 0.000 claims description 3
- 108010007622 LDL Lipoproteins Proteins 0.000 claims description 3
- XSQUKJJJFZCRTK-UHFFFAOYSA-N Urea Chemical compound NC(N)=O XSQUKJJJFZCRTK-UHFFFAOYSA-N 0.000 claims description 3
- LEHOTFFKMJEONL-UHFFFAOYSA-N Uric Acid Chemical compound N1C(=O)NC(=O)C2=C1NC(=O)N2 LEHOTFFKMJEONL-UHFFFAOYSA-N 0.000 claims description 3
- TVWHNULVHGKJHS-UHFFFAOYSA-N Uric acid Natural products N1C(=O)NC(=O)C2NC(=O)NC21 TVWHNULVHGKJHS-UHFFFAOYSA-N 0.000 claims description 3
- 230000001154 acute effect Effects 0.000 claims description 3
- 201000011510 cancer Diseases 0.000 claims description 3
- 239000004202 carbamide Substances 0.000 claims description 3
- 235000012000 cholesterol Nutrition 0.000 claims description 3
- 229940109239 creatinine Drugs 0.000 claims description 3
- 238000000502 dialysis Methods 0.000 claims description 3
- 239000008103 glucose Substances 0.000 claims description 3
- 238000005259 measurement Methods 0.000 claims description 3
- 238000012360 testing method Methods 0.000 claims description 3
- 150000003626 triacylglycerols Chemical class 0.000 claims description 3
- 229940116269 uric acid Drugs 0.000 claims description 3
- 206010028980 Neoplasm Diseases 0.000 claims description 2
- 208000015181 infectious disease Diseases 0.000 claims description 2
- 230000036210 malignancy Effects 0.000 claims description 2
- 239000000758 substrate Substances 0.000 claims 1
- 238000003745 diagnosis Methods 0.000 description 10
- 208000001072 type 2 diabetes mellitus Diseases 0.000 description 8
- 101710194425 Dehydrogenase/reductase SDR family member 4 Proteins 0.000 description 6
- 241001504424 Zosteropidae Species 0.000 description 6
- 238000011862 kidney biopsy Methods 0.000 description 6
- 208000020832 chronic kidney disease Diseases 0.000 description 5
- 210000000795 conjunctiva Anatomy 0.000 description 5
- 230000009278 visceral effect Effects 0.000 description 5
- 210000003786 sclera Anatomy 0.000 description 4
- 208000011580 syndromic disease Diseases 0.000 description 4
- 206010012689 Diabetic retinopathy Diseases 0.000 description 3
- 230000007812 deficiency Effects 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 239000003814 drug Substances 0.000 description 3
- 230000001575 pathological effect Effects 0.000 description 3
- 238000010200 validation analysis Methods 0.000 description 3
- 206010062717 Increased upper airway secretion Diseases 0.000 description 2
- 206010025421 Macule Diseases 0.000 description 2
- 206010067482 No adverse event Diseases 0.000 description 2
- 238000009825 accumulation Methods 0.000 description 2
- 238000009792 diffusion process Methods 0.000 description 2
- 230000036040 emmetropia Effects 0.000 description 2
- PCHJSUWPFVWCPO-UHFFFAOYSA-N gold Chemical compound [Au] PCHJSUWPFVWCPO-UHFFFAOYSA-N 0.000 description 2
- 230000001788 irregular Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000000877 morphologic effect Effects 0.000 description 2
- 238000010827 pathological analysis Methods 0.000 description 2
- 208000026435 phlegm Diseases 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 206010010741 Conjunctivitis Diseases 0.000 description 1
- 102000012192 Cystatin C Human genes 0.000 description 1
- 108010061642 Cystatin C Proteins 0.000 description 1
- 206010054044 Diabetic microangiopathy Diseases 0.000 description 1
- 108060006698 EGF receptor Proteins 0.000 description 1
- 206010039705 Scleritis Diseases 0.000 description 1
- 241001408665 Timandra griseata Species 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 238000003556 assay Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000001574 biopsy Methods 0.000 description 1
- 210000000988 bone and bone Anatomy 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000012512 characterization method Methods 0.000 description 1
- 230000002596 correlated effect Effects 0.000 description 1
- 230000000875 corresponding effect Effects 0.000 description 1
- 201000009101 diabetic angiopathy Diseases 0.000 description 1
- 238000002405 diagnostic procedure Methods 0.000 description 1
- 238000003748 differential diagnosis Methods 0.000 description 1
- 239000006185 dispersion Substances 0.000 description 1
- -1 egffr Proteins 0.000 description 1
- 208000028208 end stage renal disease Diseases 0.000 description 1
- 201000000523 end stage renal failure Diseases 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- GNBHRKFJIUUOQI-UHFFFAOYSA-N fluorescein Chemical compound O1C(=O)C2=CC=CC=C2C21C1=CC=C(O)C=C1OC1=CC(O)=CC=C21 GNBHRKFJIUUOQI-UHFFFAOYSA-N 0.000 description 1
- 210000000232 gallbladder Anatomy 0.000 description 1
- 230000036541 health Effects 0.000 description 1
- 230000020169 heat generation Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 208000017169 kidney disease Diseases 0.000 description 1
- 235000013336 milk Nutrition 0.000 description 1
- 239000008267 milk Substances 0.000 description 1
- 210000004080 milk Anatomy 0.000 description 1
- 150000002825 nitriles Chemical class 0.000 description 1
- 230000008520 organization Effects 0.000 description 1
- 230000007170 pathology Effects 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000004393 prognosis Methods 0.000 description 1
- 238000002601 radiography Methods 0.000 description 1
- 210000001525 retina Anatomy 0.000 description 1
- 230000009885 systemic effect Effects 0.000 description 1
- 210000004291 uterus Anatomy 0.000 description 1
- 210000003462 vein Anatomy 0.000 description 1
- 210000001835 viscera Anatomy 0.000 description 1
Classifications
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
- G16H50/20—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for computer-aided diagnosis, e.g. based on medical expert systems
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0012—Biomedical image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/77—Processing image or video features in feature spaces; using data integration or data reduction, e.g. principal component analysis [PCA] or independent component analysis [ICA] or self-organising maps [SOM]; Blind source separation
- G06V10/774—Generating sets of training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H10/00—ICT specially adapted for the handling or processing of patient-related medical or healthcare data
- G16H10/60—ICT specially adapted for the handling or processing of patient-related medical or healthcare data for patient-specific data, e.g. for electronic patient records
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30004—Biomedical image processing
- G06T2207/30041—Eye; Retina; Ophthalmic
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30004—Biomedical image processing
- G06T2207/30084—Kidney; Renal
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
- Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- General Health & Medical Sciences (AREA)
- Theoretical Computer Science (AREA)
- Public Health (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Databases & Information Systems (AREA)
- Multimedia (AREA)
- Biomedical Technology (AREA)
- Epidemiology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Primary Health Care (AREA)
- Pathology (AREA)
- Quality & Reliability (AREA)
- Radiology & Medical Imaging (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Data Mining & Analysis (AREA)
- Artificial Intelligence (AREA)
- Computing Systems (AREA)
- Evolutionary Computation (AREA)
- Software Systems (AREA)
- Investigating Or Analysing Biological Materials (AREA)
Abstract
The invention discloses a method and a system for predicting diabetic nephropathy based on white eye characteristics, which belong to the technical field of medical care information, and the method comprises the following steps: collecting clinical data and white eye feature data to obtain a data set; building a training set according to the data set; screening modeling indexes from a training set or a data set; training the training set based on a logistic regression method to obtain a prediction model; and predicting clinical data and white eye characteristic data through the prediction model to obtain the probability of suffering from diabetic nephropathy. Establishing a prediction model by researching the association of clinical data and white eye characteristic data with diabetic nephropathy; the probability of diabetic nephropathy is predicted through the prediction model, noninvasive prediction is realized, simplicity is realized, and the prediction result is reliable.
Description
Technical Field
The invention relates to the technical field of medical care information, in particular to a method and a system for predicting diabetic nephropathy based on white eye characteristics.
Background
Diabetic retinopathy (Diabetic retinopathy, DR) and diabetic nephropathy (Diabetic nephropathy, DN) are diabetic microangiopathy. The severity of DR is a risk factor for end stage renal disease in DN patients diagnosed by renal biopsy, and thus the severity of DR may be a powerful tool for predicting the course of DN clinical disease. Analysis of a DR predicted DN indicated that DR was useful in diagnosing or screening patients with Type 2 diabetes (Type 2 diabetes mellitus,T2DM) combined with kidney disease for a Type DN pathology from a kidney biopsy.
At present, DR is mainly diagnosed through fundus photography or invasive fluorescein radiography, in some cases, an ophthalmologist is required to perform professional evaluation, and diagnosis of DR cannot be achieved in a basic place where medical conditions fall behind, so that differential diagnosis of DN cannot be assisted on the basis of the diagnosis.
Since DN and Non-diabetic nephropathy (Non-diabetic renal disease, NDRD) are different in treatment and prognosis, it is important to distinguish DN from NDRD. Renal biopsy is a gold standard for diagnosis of DN, but is an invasive procedure because of its being an invasive procedure.
The sclera and bulbar conjunctiva are more superficial in structure and easier to view than the retina. Previous studies have acquired scleral images of diabetic patients and healthy subjects and found that the curvature of the large blood vessels on the bulbar conjunctiva of diabetic patients is significantly smaller than in non-diabetic groups. The scholars found that the conjunctival vasculopathy of Type 2 diabetes (Type 2 diabetes mellitus,T2DM) was earlier than DR. Furthermore, conjunctival vessel width and tortuosity are positively correlated with increased DR severity.
The white eye diagnosis is a diagnostic method in the diagnosis of the eye in traditional Chinese medicine, which judges systemic diseases by observing changes in the morphology, color, etc. of the sclera and bulbar conjunctiva vessels. The morphological characteristics, blood vessel color characteristics and visceral location characteristics of the white eyes can be obtained by observing the eye condition information of the sclera and the bulbar conjunctiva through the examination.
Therefore, the research of conjunctival blood vessels provides a new idea for noninvasive diagnosis of eyes in traditional Chinese medicine. It is therefore an important research direction to find a way to predict or identify DN and NDRD noninvasively based on the white eye characteristics.
Disclosure of Invention
Aiming at the technical problems in the prior art, the invention provides a method and a system for predicting diabetic nephropathy based on the characteristics of the eyes, and a prediction model of the diabetic nephropathy is established based on the characteristics of the eyes, so that the diabetic nephropathy is predicted, and the method and the system are simple, reliable and noninvasive.
The invention discloses a method for predicting diabetic nephropathy based on white eye characteristics, which comprises the following steps: collecting clinical data and white eye feature data to obtain a data set; building a training set according to the data set; screening modeling indexes from a training set or a data set; training the training set based on a logistic regression method to obtain a prediction model; and predicting clinical data and white eye characteristic data through the prediction model to obtain the probability of suffering from diabetic nephropathy.
Preferably, the clinical data includes: sex, age, BMI, mean arterial pressure, diabetes course, glycosylated hemoglobin, fasting blood glucose, 24 hour urine protein quantification, hemoglobin, fibrinogen, blood urea, blood creatinine, blood uric acid, gfr, total cholesterol, triglycerides, high density lipoproteins, and low density lipoproteins;
the indexes of the white eye characteristics include: mounds, spots, foggy, blood vessel dark pink, blood vessel dark red, blood vessel light red, spleen and stomach zone, liver and gall zone, kidney and bladder zone, lung and large intestine zone, heart and small intestine zone.
Preferably, the method for screening modeling indexes comprises the following steps:
screening out a first index with a P value smaller than a first threshold value based on the inter-group comparison of the diabetic nephropathy group and the non-diabetic nephropathy group in the training set;
and screening modeling indexes from the first indexes through multiple collinearity verification.
Preferably, the modeling index includes: the detection value of hills, age, diabetes course, hemoglobin and fibrinogen.
Preferably, the predictive model is expressed as:
；
wherein,P(DN)expressed as the probability of being predicted as diabetic nephropathy,Qthe detected value represented as a hill is,Yexpressed as the age of the person,D1is expressed as the course of the disease of diabetes,Hexpressed as a measurement value of hemoglobin,FPexpressed as a fibrinogen test value,Arepresented as a constant value, the value of which is,and->Respectively, weight coefficients.
Preferably, the specific prediction model is:
；
preferably, predicting whether the patient suffers from diabetic nephropathy based on the probability and a preset second threshold;
the invention also includes rules for excluding clinical data and white-eye feature data, the rules for excluding including any of the following data:
data for pregnant or lactating women; data from kidney transplant or dialysis patients; data of malignancy patients; data from patients with acute severe infectious disease; data of ophthalmic diseases affecting the observation of white eye characteristics.
The invention also provides a system for realizing the method, which comprises a prediction module, wherein the prediction module is used for predicting clinical data and white eye characteristic data through the prediction model to obtain the probability of suffering from diabetic nephropathy.
Preferably, the system further comprises an acquisition module, a preprocessing module and a training module,
the acquisition module is used for acquiring clinical data and white eye characteristic data to obtain a data set;
the preprocessing module is used for establishing a training set according to the data set; screening modeling indexes from a training set or a data set;
the training module is used for training the training set based on a logistic regression method to obtain a prediction model;
the prediction module is further configured to predict whether the patient has diabetic nephropathy according to the probability and a preset second threshold.
The invention also provides a device comprising a memory and a processor, wherein the memory is used for storing instructions for executing the method for predicting diabetic nephropathy based on the white eye characteristics; the processor is configured to execute the instructions.
Compared with the prior art, the invention has the beneficial effects that: establishing a prediction model by researching the association of clinical data and white eye characteristic data with diabetic nephropathy; the probability of diabetic nephropathy is predicted through the prediction model, noninvasive prediction is realized, simplicity is realized, and the prediction result is reliable.
Drawings
FIG. 1 is a flow chart of a method of the invention for predicting diabetic nephropathy based on white eye characteristics;
FIG. 2A is a sequence diagram of an operation for observing an eye image of a subject;
FIG. 2B is a diagram showing the visceral location distribution of the surface of the eye;
FIG. 3 is a ROC graph of predictive model training;
FIG. 4 is a ROC graph of predictive model verification;
fig. 5 is a logical block diagram of a system for predicting diabetic nephropathy based on white eye features of the present invention.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the embodiments of the present invention more apparent, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention, and it is apparent that the described embodiments are some embodiments of the present invention, but not all embodiments of the present invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
The invention is described in further detail below with reference to the attached drawing figures:
a method for predicting diabetic nephropathy based on white eye characteristics, as shown in fig. 1, comprising the steps of:
step 101: clinical data and white eye feature data are collected to obtain a data set.
Step 102: from the dataset, a training set is established.
Step 103: modeling metrics are screened from a training set or dataset.
Step 104: training the training set based on a logistic regression method to obtain a prediction model.
Step 105: and predicting clinical data and white eye characteristic data through the prediction model to obtain the probability of suffering from diabetic nephropathy.
Step 106: and verifying the prediction model by using a verification set.
Establishing a prediction model by researching the association of clinical data and white eye characteristic data with diabetic nephropathy; the probability of diabetic nephropathy is predicted through the prediction model, noninvasive prediction is realized, simplicity is realized, and the prediction result is reliable.
And (3) data acquisition: the following inclusion and exclusion rules are considered in the data acquisition. The inclusion criteria were: (1) compliance with a T2DM diagnosis; (2) Conforming to chronic kidney disease (Chronic Kidney Disease, CKD) diagnosis; (3) the age is more than or equal to 18 years old; (4) clear pathological diagnosis of kidney puncture biopsy; (5) a specialized ophthalmic physician diagnoses fundus results. The exclusion criteria were: (1) pregnant or lactating women; (2) kidney transplant or dialysis patient; (3) a patient with malignant tumor; (4) patients with acute severe infectious diseases; (5) Conjunctivitis, scleritis, etc. affect patients who observe eye-specific ophthalmic diseases.
Among them, T2DM was diagnosed according to the 1998 world health organization's standard. CKD was diagnosed according to kdaigo guidelines in 2012. DR was diagnosed based on early treatment studies of diabetic retinopathy. DN all pathological results are diagnosed by independent reading of two independent renieratcs with abundant experience according to DN pathological standard published by 2010 International kidney pathological society.
General indicators of clinical data include: sex, age, BMI, mean arterial pressure, course of diabetes; the clinical indexes include: glycosylated hemoglobin, fasting blood glucose, 24-hour urine protein quantification, hemoglobin, fibrinogen, blood urea, blood creatinine, blood uric acid, eGFR, total cholesterol, triglycerides, high density lipoproteins, and low density lipoproteins.
Referring to fig. 2A, the eye image of the subject was observed according to the following procedure: left eye emmetropia, left eye left view, left eye right view, left eye up view, left eye down view, right eye emmetropia, right eye left view, right eye right view, right eye up view, right eye down view.
The white eye adopted by the invention has the following characteristics: (1) white eye morphology: hills, spots, fog. (2) white blood vessel color characterization: blood vessels were dark pink, blood vessels were dark red, blood vessels were pale red, blood vessels were pink. (3) visceral locations on the surface of the white eyes: spleen and stomach (R1+R2+R15), liver and gall (R12+R13+R14), kidney and bladder (R8+R9+R10), lung and large intestine (R3+R7), heart and small intestine (R4+R5). The visceral locations on the surface of the eye are shown in FIG. 2B. During the detection process, each white eye feature appears once, and the integral is 1 minute.
Wherein, the hills refer to the opaque larger ridges with diameters above 0.2cm and a circular, elliptic or irregular shape on the surface of the high Yu Bai nitrile. The patch refers to a colored patch which is round, oval or various irregular forms and does not bulge on the surface of the white eye. The dots refer to circular color dots that do not protrude from the white surface. The fog diffusion refers to a color dip state which is emitted from the fornix part and is spread to the whole white eyes in a fog dispersion shape and does not bulge on the surface of the white eyes. The pattern of hillock with accumulation of damp-phlegm, the pattern of macula with accumulation of damp-phlegm or blood stasis, the pattern of macula with stagnation of qi and blood stasis, and the pattern of fog with diffusion of wind. The dark pink blood vessels are the syndromes of blood deficiency and blood stasis, the dark red blood vessels are the syndromes of blood stasis and excessive heat, the pale blood vessels are the syndromes of qi deficiency, and the pink blood vessels are the syndromes of blood deficiency and heat generation.
In FIG. 2B, the visceral location distribution R1 of the white eyes: stomach, R2: spleen, R3: large intestine, R4: small intestine, R5: heart, R6: milk, R7: lung, R8: kidney, R9: bladder, R10: kidney, R11: female uterus, male outer kidney, bone lumbo-sacral leg and foot and corresponding medulla, R12: liver, R13: bladder, R14: liver, R15: spleen.
Finally, 212 patients were included as training sets, 103 of the DN groups and 109 of the NDRD groups. The clinical index and the parameters of the clinic are shown in table 1.
TABLE 1
Index screening:
based on the inter-group comparison of the parameters of the clinic equipment and the clinical indexes, the parameters of the clinic equipment with P less than 0.2 are screened out: mounds, plaques, foggies, dark pink veins, spleen and stomach regions, liver and gall bladder regions, kidney and bladder regions, heart and small intestine regions; clinical index: age, BMI, mean arterial pressure, diabetes course, 24h urine protein quantification, hemoglobin, fibrinogen, egffr, cystatin C and high density lipoprotein, and after multiple co-linearity testing of these indices, the following modeling index was selected: hills, age, diabetes course, hemoglobin, and fibrinogen.
Model training: and training a prediction model based on a multi-factor logistic regression algorithm. The logistic regression analysis results are shown in table 2, wherein hills (or=1.409; 95% ci=1.109-1.789; p=0.005), ages (or=0.930; 95% ci=0.893-0.968; p < 0.001), course of diabetes (or=1.014; 95% ci=1.008-1.019; p < 0.001), haemoglobin (or=0.975; 95% ci=0.959-0.991; p=0.002), fibrinogen (or=1.418; 95% ci=1.114-1.804; p=0.005) are independent factors of DN.
TABLE 2
The trained diabetic nephropathy prediction model is:
；
wherein,P(DN)expressed as the probability of being predicted as diabetic nephropathy,Qthe detection value being expressed as a hill, i.e. the occurrence of a hillIs used for the frequency of (a),Yexpressed as the age of the person,D1is expressed as the course of the disease of diabetes,Hexpressed as a measurement value of hemoglobin,FPexpressed as fibrinogen assay. It should be noted that the trained predictive model systems differ slightly from one training set to another. The predictive model may be expressed generally as:
；
wherein,Arepresented as a constant value, the value of which is,and->Respectively, weight coefficients.
FIG. 3 shows ROC curves plotted against the diagnostic probability of the predictive model with an area under the curve (AUC) of 0.845, SE of 0.026, 95% CI of 0.793-0.896 and P < 0.001.
And (3) verifying a prediction model: according to the inclusion criteria and exclusion criteria described above, 33 patients were included as a validation set. The comparison between the clinical index of the validation set and the parameters of the diagnostic unit is shown in Table 3.
TABLE 3 Table 3
Judging as DN if the predicted result P (DN) is greater than or equal to a threshold value, if the predicted result P (DN) is more than or equal to 0.5; and when the prediction result is smaller than the threshold value, judging that the NDRD is the NDRD.
Each data in the verification set was substituted into the prediction model, and the P (DN) value was calculated, with the calculation results shown in table 4.
TABLE 4 Table 4
The misjudgment of the pathological diagnosis results of the validation set is shown in table 5, the ROC curve is shown in fig. 4, the area under the curve (AUC) is 0.721, se is 0.094, 95% ci is 0.536-0.906, and p=0.034.
TABLE 5
Compared with the existing kidney biopsy, the prediction model and the method for predicting the diabetic nephropathy by applying the prediction model have the advantages of no wound, rapidness and high efficiency, no adverse reaction and good reliability.
The invention also provides a system for realizing the method, as shown in fig. 5, which comprises a prediction module 5, wherein the prediction module 5 is used for predicting clinical data and white eye characteristic data through the prediction model to obtain the probability of diabetic nephropathy.
The system also comprises an acquisition module 1, a preprocessing module 2 and a training module 3,
the acquisition module 1 is used for acquiring clinical data and white eye characteristic data to obtain a data set;
the preprocessing module 2 is used for establishing a training set according to the data set; screening modeling indexes from a training set or a data set;
the training module 3 is used for training the training set based on a logistic regression method to obtain a prediction model;
the prediction module 5 is further configured to predict whether the patient has diabetic nephropathy according to the probability and a preset second threshold.
The present invention also provides an apparatus comprising a memory for storing instructions for performing the method of predicting diabetic nephropathy based on white eye characteristics.
Compared with gold standard kidney biopsy, the invention observes the eye condition information of sclera and bulbar conjunctiva through traditional Chinese medicine eye diagnosis, obtains morphological characteristics, blood vein color characteristics and viscera position characteristics of the white eyes, and quantifies white eye parameters; training a prediction model according to the quantized parameters; has the advantages of no wound, high speed and high efficiency, and no adverse reaction.
It should be noted that the predicted outcome of the present invention does not represent the final diagnostic outcome, but is merely used to provide a reference for further diagnosis and treatment of the patient, e.g. the patient may be advised to take a renal biopsy if the predicted probability is above a certain threshold.
The above is only a preferred embodiment of the present invention, and is not intended to limit the present invention, but various modifications and variations can be made to the present invention by those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (6)
1. A method for predicting diabetic nephropathy based on white eye characteristics, comprising the steps of:
collecting clinical data and white eye feature data to obtain a data set;
building a training set according to the data set;
screening modeling indexes from a training set or a data set, wherein the modeling indexes comprise: detection value of hills, age, diabetes course, hemoglobin and fibrinogen;
training the training set based on a logistic regression method to obtain a prediction model;
predicting clinical data and white eye feature data through the prediction model to obtain the probability of suffering from diabetic nephropathy;
wherein the clinical data comprises the following indexes: sex, age, BMI, mean arterial pressure, diabetes course, glycosylated hemoglobin, fasting blood glucose, 24 hour urine protein quantification, hemoglobin, fibrinogen, blood urea, blood creatinine, blood uric acid, gfr, total cholesterol, triglycerides, high density lipoproteins, and low density lipoproteins;
the indexes of the white eye characteristics include: mounds, spots, foggy, blood vessel dark pink, blood vessel dark red, blood vessel light red, spleen and stomach zone, liver and gall zone, kidney and bladder zone, lung and large intestine zone, heart and small intestine zone;
the method for screening modeling indexes comprises the following steps:
screening out a first index with a P value smaller than a first threshold value based on the inter-group comparison of the diabetic nephropathy group and the non-diabetic nephropathy group in the training set;
screening modeling indexes from the first indexes through multiple collinearity verification;
the predictive model is expressed as:
；
wherein,P(DN)expressed as the probability of being predicted as diabetic nephropathy,Qthe detected value represented as a hill is,Yexpressed as the age of the person,D1is expressed as the course of the disease of diabetes,Hexpressed as a measurement value of hemoglobin,FPexpressed as a fibrinogen test value,Arepresented as a constant value, the value of which is,and->Respectively, weight coefficients.
2. The method of claim 1, wherein the step of determining the position of the substrate comprises,
predicting whether the patient has diabetic nephropathy based on the probability and a preset second threshold.
3. The method of claim 1, wherein the specific predictive model is:
。
4. the method of claim 1, further comprising an exclusion rule for clinical data and white-eye feature data, the exclusion rule comprising any of the following data:
data for pregnant or lactating women; data from kidney transplant or dialysis patients; data of malignancy patients; data from patients with acute severe infectious disease; data of ophthalmic diseases affecting the observation of white eye characteristics.
5. A system for predicting diabetic nephropathy based on white-eye features, comprising an acquisition module, a preprocessing module, a training module, and a prediction module for implementing the method of any one of claims 1-4,
the acquisition module is used for acquiring clinical data and white eye characteristic data to obtain a data set;
the preprocessing module is used for establishing a training set according to the data set; screening modeling indexes from a training set or a data set;
the training module is used for training the training set based on a logistic regression method to obtain a prediction model;
the prediction module is used for predicting clinical data and white eye characteristic data through the prediction model to obtain the probability of diabetic nephropathy.
6. The system of claim 5, wherein the prediction module is further configured to predict whether the patient has diabetic nephropathy based on the probability and a preset second threshold.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311174475.8A CN116913508B (en) | 2023-09-13 | 2023-09-13 | Method and system for predicting diabetic nephropathy based on white eye characteristics |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311174475.8A CN116913508B (en) | 2023-09-13 | 2023-09-13 | Method and system for predicting diabetic nephropathy based on white eye characteristics |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116913508A CN116913508A (en) | 2023-10-20 |
| CN116913508B true CN116913508B (en) | 2023-12-12 |
Family
ID=88360675
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311174475.8A Active CN116913508B (en) | 2023-09-13 | 2023-09-13 | Method and system for predicting diabetic nephropathy based on white eye characteristics |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116913508B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117727459A (en) * | 2024-02-18 | 2024-03-19 | 中国人民解放军总医院第一医学中心 | Dialysis opportunity prediction method and system for chronic kidney disease 5-phase combined type 2 diabetes mellitus |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101874824A (en) * | 2009-05-01 | 2010-11-03 | 常州高新技术产业开发区三维工业技术研究所有限公司 | Active extract containing trilobatin and application thereof |
| CN112513999A (en) * | 2017-10-13 | 2021-03-16 | 人工智能技术公司 | Deep learning based diagnosis and referral of ophthalmic diseases and conditions |
| CN113273959A (en) * | 2021-07-19 | 2021-08-20 | 中山大学中山眼科中心 | Portable diabetic retinopathy diagnosis and treatment instrument |
| CN113889267A (en) * | 2021-11-15 | 2022-01-04 | 博奥生物集团有限公司 | Method for constructing diabetes diagnosis model based on eye image recognition and electronic equipment |
| CN114038564A (en) * | 2021-11-11 | 2022-02-11 | 天津医科大学朱宪彝纪念医院(天津医科大学代谢病医院、天津代谢病防治中心) | Noninvasive risk prediction method for diabetes |
| CN116635889A (en) * | 2020-11-02 | 2023-08-22 | 谷歌有限责任公司 | Machine learning to detect disease from external anterior eye images |
-
2023
- 2023-09-13 CN CN202311174475.8A patent/CN116913508B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101874824A (en) * | 2009-05-01 | 2010-11-03 | 常州高新技术产业开发区三维工业技术研究所有限公司 | Active extract containing trilobatin and application thereof |
| CN112513999A (en) * | 2017-10-13 | 2021-03-16 | 人工智能技术公司 | Deep learning based diagnosis and referral of ophthalmic diseases and conditions |
| CN116635889A (en) * | 2020-11-02 | 2023-08-22 | 谷歌有限责任公司 | Machine learning to detect disease from external anterior eye images |
| CN113273959A (en) * | 2021-07-19 | 2021-08-20 | 中山大学中山眼科中心 | Portable diabetic retinopathy diagnosis and treatment instrument |
| CN114038564A (en) * | 2021-11-11 | 2022-02-11 | 天津医科大学朱宪彝纪念医院(天津医科大学代谢病医院、天津代谢病防治中心) | Noninvasive risk prediction method for diabetes |
| CN113889267A (en) * | 2021-11-15 | 2022-01-04 | 博奥生物集团有限公司 | Method for constructing diabetes diagnosis model based on eye image recognition and electronic equipment |
Non-Patent Citations (3)
| Title |
|---|
| Renoprotective Effect of Yiqi Yangyin Huayu Tongluo Formula against Diabetic Nephropathy in Diabetic Rats;Feng-li Wang;《Evidence-Based Complementary and Alternative Medicine》;全文 * |
| 眼部相关全身疾病的人工智能诊断;李金城等;《眼科学报》;第37卷(第3期);第222-229页 * |
| 糖尿病肾病中医证候的现代研究;王倩等;《世界中医药》;第13卷(第6期);第1320-1324页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116913508A (en) | 2023-10-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Pan et al. | Multi-label classification of retinal lesions in diabetic retinopathy for automatic analysis of fundus fluorescein angiography based on deep learning | |
| Zhang et al. | A survey on computer aided diagnosis for ocular diseases | |
| Garway-Heath et al. | Quantitative evaluation of the optic nerve head in early glaucoma | |
| CN116913508B (en) | Method and system for predicting diabetic nephropathy based on white eye characteristics | |
| Sandhu et al. | Automated diagnosis and grading of diabetic retinopathy using optical coherence tomography | |
| Kauppi | Eye fundus image analysis for automatic detection of diabetic retinopathy | |
| JP2006507577A (en) | Medical image retrieval method and system | |
| US20150250430A1 (en) | Noninvasive method for determining the presence of systemic hypertension in a subject | |
| KR102071774B1 (en) | Method for predicting cardio-cerebrovascular disease using eye image | |
| Kumar et al. | Automated detection of eye related diseases using digital image processing | |
| CN116913524B (en) | Method and system for predicting diabetic nephropathy based on retinal vascular imaging | |
| Zheng et al. | Research on an intelligent lightweight-assisted pterygium diagnosis model based on anterior segment images | |
| Samagaio et al. | Automatic segmentation of diffuse retinal thickening edemas using Optical Coherence Tomography images | |
| CN111096727B (en) | Method and system for detecting pregnancy-induced hypertension, electronic device and storage medium | |
| KR20200011530A (en) | Method for predicting cardio-cerebrovascular disease using eye image | |
| Feng et al. | Macular vascular geometry changes with sex and age in healthy subjects: a fundus photography study | |
| Tugashetti et al. | Design of algorithms for diagnosis of primary glaucoma through estimation of CDR in different types of fundus images using IP techniques | |
| Kook et al. | Scanning laser polarimetry using variable corneal compensation in the detection of glaucoma with localized visual field defects | |
| CN116313037A (en) | Diabetes prediction method and system based on multi-modal data | |
| CN115768359A (en) | Blood abnormality prediction device, blood abnormality prediction method, and program | |
| Agarwal et al. | Non-invasive techniques for the screening of diabetic retinopathy | |
| Betzler et al. | Artificial Intelligence in Predicting Systemic Disease from Ocular Imaging: Teleophthalmology and Digital Health: A Practical Guide to Applications | |
| Ghebrechristos et al. | RetiNet—feature extractor for learning patterns of diabetic retinopathy and age-related macular degeneration from publicly available datasets | |
| US20230154004A1 (en) | Retinal vascular stress test for diagnosis of vision-impairing diseases | |
| CN115083005B (en) | ROP image classification system and method based on deep learning |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |